Light emitting diodes (“LED”) are ubiquitous in electronics. They are used in digital displays, lighting systems, computers and televisions, cellular telephones and a variety of other devices. In an LED, as in a traditional diode, extra electrons move from an N-type semiconductor to electron holes in a P-type semiconductor. In an LED, however, photons are released to produce light during this process. For many applications it is desirable to collect as much light as possible from the LED and distribute it into a desired cone angle.
Many conventional LED devices use a dome spherical or aspheric lens formed around the LED. Generally, the distance from the lens to the dome controls the emission cone. The T-1¾, T-5 mm or variations thereof are examples of dome lens LEDs. However, there are several drawbacks to this design. First, typical domes can only collect an f/1 acceptance angle of the LED die. Hence, photons emitted greater than this angle are either trapped within the dome due to total internal reflection (“TIR”) or emitted out the edge of the dome at a non-usable angle. Next, the distribution of the light is highly dependent on the accuracy of the alignment between the chip and the dome. Therefore, far field and near field distributions are often sacrificed. Third, there can be significant non-uniformities between the near-field and far field distribution. Lastly, the distribution itself is not spatially uniform.
Another conventional scheme is to place a larger dome on top of the LED. Though this method does allow most if not all of the energy to get out, there are several significant drawbacks for practical applications. First, the emission cone angle is typically greater than 180 degrees. Though light is no longer trapped, energy is emitted to an angle greater than the original angle of the LED. Mechanical housings and such can vignette, scatter and absorb the light at the larger angles. Moreover, since most secondary optical systems only collect an f/1 cone (a cone having a half angle of approximately 30 degrees or less), much of the light is lost. Thirdly, since the dome is much larger than the LED die, the distribution is over a much larger area than necessary. This translates into a lower power density (or irradiance) when the light is focused.
Another solution is to place a TIR lens over the typical dome lens to collect all of the emitted energy and direct it into a smaller cone. This adds complexity to the system and only addresses the problem of getting more light into a narrower cone angle. These systems also do not address conservation of brightness of the source, creating a uniform pattern and maintaining the uniformity far field as well as near field. Also, adding such a TIR lens increases the size and cost of a lighting package as much as tenfold, rendering this solution impractical for nearly all LED applications in electronics and portable devices. Other systems utilize elaborate TIR lens, reflective collectors and condenser lens systems. While some reflective systems that re-image the LED from a dome can maintain the radiance (e.g., an ellipsoid where the LED is at one foci and the image is at the other foci), these systems are impractical for many applications.